Introduction

Existing studies of international relations in East Asia from the seventh to the ninth centuries are mainly concerned with the turmoil between the Silla, Baekje, and Goguryeo or the impact of the Tang Dynasty on the three Kingdoms, either in terms of their involvement in wars or who they were with. But how did Japan view foreign international relations during this period? No (2012) and Owada (2012) focused on the history of unification wars between three countries on the Korean Peninsula. However, they did not address how Japan views external relations. In addition, despite some scholarly discussion, the time point does not focus on the first half of the seventh century. Suzuki (2011) discussed Japan’s relations with East Asian countries in the second half of the seventh century and argued that the first half of the seventh century was a period of great international tension, reflected in the expansion of the Tang Dynasty’s power and attempts to invade the Korean peninsula. Chen (2012) also drew attention to Tang–Japanese relations from the Battle of Baekgang to the beginning of the eighth century, pointing out that Tang–Japanese relations centered around the Korean peninsula. He highlighted that the Japanese army intervened in Baekje’s foreign wars but returned in failure. As such, Japan no longer intervened directly in Korean affairs after the war. However, the researchers did not discuss how Japan viewed its foreign relations in the first half of the seventh century. To respond to this critical void, Kitou (1976) made the case that the relations of the states from the 630 s to the 670 s were based on the Tang’s foreign policy, which was based on a political and categorical approach. During this era, Japan was separated from the global political stage dominated by the Tang Dynasty (Yamao, 1977). In addition, Shen’s (1994) research covering the 7th and 8th centuries examined the Tang Dynasty’s perspective on the Korean Peninsula and its varying stances towards Silla and Japan. However, it did not delve into Japan’s perceptions of its foreign diplomatic relations then.

While there have been more scholarly ontological studies of Japanese fortifications in the academic world, especially in terms of the siting and construction of ancient mountain fortresses (see Takahashi, 1972), only a few draw attention to the techniques of fortification and the objects left behind in ancient mountain fortresses (Historical PKC, 2016). Some scholars have also analyzed the relations between East Asian countries during this period from the history of cultural exchanges between China and Japan. For instance, Kimiya (1955) investigated Japan’s evolving circumstances on the Korean Peninsula by analyzing the paths taken by the Kentoshi envoys to the Tang Dynasty. During the first half of the 7th century, Japan opted for the northern land route to Tang due to improving relations with Baekje and Goguryeo. However, in the latter half of the century, amid the destruction of both nations and strained ties with Silla, Japan shifted to the southern sea route. Fujiie (1977) posited that sending envoys to the Tang Dynasty in the earlier 7th century served a dual purpose. It aimed to acquire knowledge about the Tang Dynasty’s system and culture while also being influenced by the complex interplay of China and Japan’s competition on the Korean Peninsula. This rivalry was exacerbated by Japan’s ties to Baekje and Goguryeo, particularly following the Tang Dynasty’s military expeditions in Goguryeo. Recognizing such an issue, Ge (2007) delved into the interactions of Tang Dynasty envoys in the late 8th and early 9th centuries, shedding light on the intricate Tang–Japanese relations during that period.

While scholars have extensively researched Sino-Japanese and East Asian relations during the Tang Dynasty, it constitutes a critical void regarding how Japan’s construction of fortifications reflects the changing nature of its international relationships with ancient China. As such, this paper seeks to fill this gap by examining Japan’s defensive fortifications, including their background, reasons, timing, and development. It will also delve into the scale of construction at Itojo and Onojo, suggesting that Japan’s sense of external tensions predated the battle of Baekgang. From this perspective, this paper will unveil the changes in the international situation and Japan’s perception of the international relations in East Asia that were hidden behind the construction of fortifications.

The battle of Baekgang and Mizuki

In the first century B.C., the Korean peninsula witnessed the rise of three kingdoms: Goguryeo, Silla, and Baekje. By the 7th century, during the establishment of the Tang Dynasty in China, Silla came under attack from both Goguryeo and Baekje. Struggling with the ongoing warfare, Silla sought assistance from the Tang Dynasty. This led to the involvement of two prominent military forces in East Asia: the Tang Dynasty supporting Silla against Goguryeo and Baekje. In AD 654, Emperor Gaozong of the Tang Dynasty issued an order to the KentoshiFootnote 1, “Emperor Gaozong has given a seal, ordering that troops be sent to aid Silla” (Ouyang and Song, 1975). In AD 660, the Tang–Silla coalition brought about the downfall of Baekje. The Baekje general, Kwisil Poksin, sent an envoy to Japan to ask for military support and the return of Baekje Prince Buyeo Pung. Empress Saimei decreed that The Baekje has come to me in desperation and will survive salvation (Toneri, 1965, p. 347).

Against this background, the first significant battle in China and Japan’s history occurred. The two sides fought at the Ju-Ryu fortress, which, according to the Jiutangshu, was the nest of the Ju-Ryu bandits, and if Ju-Ryu was captured, the cities would fall automatically (Liu, 1975). While the Tang and Silla forces fought together on land and water, the Japanese army split their expeditionary force into two. 10,000 Japanese troops were sent to Baekgang, while the Tang general led a fleet of 170 ships already arrayed on the Baekgang. The Japanese army was attacked by the Tang army, and “400 boats were burned, and the sea was red, smoky and burning sky” (Sima, 2011). The significantly superior Tang army achieved victory in the Battle of Baekgang, resulting in the defeat of the Japanese forces. This outcome forced the Japanese to retreat and redirect their naval resources back to Japan, thus ending any hopes of supporting the revival of Baekje. Various scholars hold differing opinions on Japan’s involvement in the Baekgang War. Some posit that the conflict served as a means to quell internal opposition to reforms (Chen, 2012; Zhang and Guo, 2006), while others argue that Japan harbored territorial ambitions over Baekje and Goguryeo (Han, 2009). Nevertheless, the predominant perspective is that Japan’s misguided confidence led to a miscalculation of the battlefield dynamics against the Tang forces, exemplified by the naval vessels employed by both sides.

In the Battle of Baekgang, the Tang army used the Mengchong and Louchuan warships, as documented in Wujingzongyao. The battle marked a turning point in East Asian history, cementing the Tang Dynasty’s power and influencing Japan’s future foreign policy decisions.

The Mengchong is a boat that uses raw bull leather to cover the back of a warship and opens the hollow of the hawser to the left and right so that a rock cannot defeat it. Crossbow windows and spear holes are at the front and back, so they will trigger an attack if the enemy approaches. This design does not rely on a large ship but prioritizes speed and takes advantage of the enemy’s unpreparednessFootnote 2 (Zeng, 2017).

The Battle of Baekgang and Mizuki marked a significant turning point in the relationship between the Tang Dynasty and Japan. In contrast to the Tang Dynasty, the Japanese possessed many ships but could not build and position them effectively. The unexpected clash with the Tang Dynasty led to a shift in focus away from pursuing victory. Instead, attention turned towards assessing the formidable presence of the old Baekje forces and the Japanese, prior to advancing northward with the entire army, as highlighted by Kim (1980). This leads to the following question: what was the position of the Japanese at this time? Pained by the power of the Tang Dynasty, the country’s foreign policy changed: relations between Japan and the Tang Dynasty were restored and strengthened with the increasing dispatch of envoys to Japan (Song, 2014). However, the dispatch of Kentoshi was closely linked to the dispatch of Tang envoys, and for the first time in the sixth year of the Tenchi era, Tang guest envoysFootnote 3 were dispatched to the Tang Dynasty. The Tang Dynasty sent positive diplomatic signals to Japan to stabilize the situation in Korea and prevent Japan from rejoining the anti-Tang forces in Baekgang (Chen, 2012), while Japan resisted such a way of sending envoys from the Tang Dynasty. After the defeat at Baekgang, Japan felt the unprecedented impact of the Tang Dynasty and realized the difference in military power between the Tang and Japan. To bridge this gap and allay fears of an imminent attack by the Tang–Silla alliance (Chen, 2012), Japan started building extensive defensive fortifications and adopted a defensive approach in its foreign policy, aiming to achieve a power balance with the Tang Dynasty. The construction of these defenses was not only to deter the Tang Dynasty but also to centralize authority within Japan. This change in foreign policy marked a considerable shift from their previous strategy of seeking closer relations with the Tang Dynasty.

During the same period, the domestic situation in Japan also underwent significant changes. In AD 645, Prince Nakano Oe initiated a major event in the history of Japan, the Isshi incident, which led to the extermination of the Suga clan, thus initially realizing the centralization of power. In the following year, Emperor Kotoku began the Taika Reforms. This includes inter alia, the imposition of administrative divisions and the system of tsu-yung-tiao, as well as direct State control over the land and the people. The purpose of the Daika Reforms was to strengthen the centralization of power. In a broader sense, the Daika Reforms lasted from AD 645 until the promulgation of the Taiho Code in AD 701, so the construction of fortifications was carried out during the period of political reforms in Japan.

While there is no doubt that Japanese fortifications were initially built to defend Japan against foreign enemies, some scholars have also explored the relationship between fortifications and the domestic situation in Japan (Kuramoto, 2010; Suzuki, 2011; Yamazaki, 2018). By way of illustration, the construction of fortifications was seen as a means of suppressing domestic local unrest and strengthening centralized power. According to Kuramoto (2010), the greatest obstacle to the Daika Reforms was the local clans that clung to their vested interests and disobeyed the central government. Therefore, by investing the power of the local clans in the construction of fortifications, the local resistance was weakened, and in this sense, the centralized power was strengthened. Suzuki (2011) argued that Japan took advantage of the Isshi incident as an opportunity for a long-term transformation of the rule of the archipelago, aiming to establish a centralized state with royal authority, which was followed by the Jinshin Rebellion of AD 672, centered on measures to control the powerful families, the bureaucracy and the legalization of rule. The building of mountain fortresses in various parts of Japan was associated with this trend and must be positioned as part of such a movement. Additionally, scholars have speculated that the construction of ancient mountain fortresses may have led to the establishment of local governmental institutions and provided a base for food storage, transport, and communication and that a series of ancient mountain fortresses served as “court centers” for the establishment of the new Ritsuryo system (Yamazaki, 2018). The history of Ehime Prefecture (1984) also pointed out that the ancient mountain fortresses may have been a stronghold for food storage, transport, and communication.

However, it has been observed that most of the pertinent views expressed by these academics are hypotheses or conjectures. While recognizing the possible influence of internal factors on urban development, our study focuses on scrutinizing historical proof. Looking through this lens, internal conditions do not offer strong validation. Hence, the effect of internal factors on urban development is viewed as a secondary aspect, with greater importance placed on the impact of external forces.

Nihongi has the following records of the construction of the castle.

In the third year of the Tenchi period, Fang (guards) and Feng (beacon fires) were set up on Tsushima Island, Iki Island, and the country of Tsukushi. And a great embankment was built in Tsukushi to store water, named MuzukiFootnote 4 (Toneri, 1965, p. 363).

In the fourth year of Tenchi, 8th month, Dalsol, Eokrye Bokryu, and Dalsol, Sabi Bokbu were sent to build Onojo and Kiijo in the country of TsukushiFootnote 5 (Toneri, 1965, p. 363).

In the sixth year of the Tenchi era, Takayasujo, Yashimajo (Yamada-gun, the country of Sanuki), and Kanetajo (the country of Tsushima) were builtFootnote 6 (Toneri, 1965, p. 367).

In the eighth year of the Tenchi era, in the winter, Takayasujo was rebuilt, and the tax on the fields in the region was collectedFootnote 7 (Toneri, 1965, p. 367).

Table 1 below shows the compiled information about the construction of Joseon-style mountain fortressesFootnote 8 based on information retrieved from the Takamatsu City Education CommitteeFootnote 9 (Historical Park Kikuchi castle・Onkosouseikan, 2019).

Table 1 Mountain fortress Information Summary Table.

In the year following the defeat of Japan, guards and beacon fires were established on the islands of Tsushima and Iki, along the northern coast of Kyushu. According to Nihongi, the Muzuki was constructed in Tsukushi, followed by Onojo and Kiijo the next year. Takayasujo was built in Kinki, Yashimajo in the Seto Inland Sea, and Kanetajo on Tsushima Island two years later. These locations served as key defensive points in Japan’s post-war strategy, which focused on establishing a coastal garrison system and a communication network using beacon fires. This led to the creation of defensive strongholds centered around Dazaifu, with multiple fortifications spanning from northern Kyushu through the Seto Inland Sea to Kinai (Zhao, 2015).

Within a few years, numerous defensive structures emerged throughout the nation. Most Japanese mountain fortresses were strategically situated in coastal regions or along rivers, such as the Seto Inland Sea. The continuous erection of fortifications reflected the apprehension and profound sense of alarm of the Japanese authorities towards external threats. Nonetheless, Japan’s diplomatic relations remained uninterrupted during this period. In the third year of Emperor Tenchi’s reign, General Liu Renyuan of Baekje dispatched Court Official Guo Wucong with a proposal letter. However, Japan refused to see the Tang ambassador because “those who have come to see us today are not ambassadors of Tianzi, but private ambassadors of the Baekje General” (Zuikei, 1995). This showed a negative diplomatic stance towards the Tang Dynasty’s positive diplomacy. During the defeat at Baekgang, Japan prioritized defending against the attacks of the Tang and Silla and building defensive fortifications, rather than restoring diplomatic relations with the Tang. In the same year, a triple defense line was set up. The Muzuki was also built in Tsukushi.

The Muzuki is a landfill structure built in AD 664. It is a large-scale embankment-like structure with a total length of 1.2 km, a width of 80 m at the base, and a height of over 13 m. Excavations have revealed a 60 m wide, 4 m deep reservoir on the Hakata side, so the general interpretation is that water was stored downstream of the Muzuki embankmentFootnote 10 (Shimatani and Hayashi, 2015).

The strategic importance of Dazaifu, a critical administrative center in ancient Japan, was not lost on its rulers. They built a series of defensive fortifications to protect the city from potential invaders, including Muzuki. To further fortify the city, a 60 m wide and 4 m deep ditch was dug on the Hakata side, making it harder for enemies to attack from that direction.

In the following year, craftsmen from Baekje were sent to Japan to build two more fortifications: Onojo, which was located north of Dazaifu and situated on Mount Shiouji, and Kiijo, which was built south of Dazaifu in Kiyama-cho, Miyaki-gun, Saga. Together with Muzuki, these fortifications formed a Rajo, a defensive ring surrounding Dazaifu.

In the same year, Tang sent Liu Degao, Chaosandafu, Xinzhousima ShangzhuguoFootnote 11, and Guo Wucong, the general of the Tang Dynasty in Baekje Kingdom, all of whom numbered two hundred and fifty-four, to Japan. On July 28th, they arrived at Tsushima, and on September 20th, they arrived at Tsukushi and presented their letter to the Emperor on September 22nd (Toneri, 1965, p. 365). The Tang Dynasty once more dispatched an official envoy to Japan, delivering the letter in September and returning in December. Japan sent an envoy to accompany the Tang envoy back to Tang. However, there is no detailed account of the discussions between the two nations nor any significant headway. It can be inferred that during this period, Japan’s primary concern was the development of defensive structures.

In the sixth year of Tenchi, on November 9th, the Tang general for Baekje, Liu Renyuan, sent the Perfect of Ungsan, in Ungjin Commandery, Sima Facong, of Shangzhuguo rank, and others to escort Ihashiki, Sakahibe no Muraji, of Lower Daisen rank, and his companions to the Governor-General’s residence in TsukushiFootnote 12 (Toneri, 1965, p. 367).

On November 13th, Sima Facong and others took their departure homewards. Hakatoko, Yuki no Muraji, of Lower Shosen rank, and Moroshi, Kasa no Omi, of Lower Daiotsu rank, were made Escort EnvoysFootnote 13 (Toneri 1965, p. 367).

It is worth noting that the Sender, the Ungjin Commandery, and the Recipient, the Tsukushi Governorate, indicate that they were in a position of equal diplomatic status. Spanning from the 9th to the 13th, a period of 5 days in total, this implies that Sima Facong’s brief visit did not result in any substantial diplomatic achievements. However, the arrival of the Tang guest envoysFootnote 14 marked a positive development in Tang–Japanese diplomacy. Despite this, tensions persisted in Japan, and that same year saw the construction of the defensive castles of Koan, Yashima, and Kaneta.

In the tenth year of Tenchi, the Tang envoys, Guo Wucong and his suite, numbering 600 persons, escorting Sataku Sonto and his companions, who numbered 1400 persons, the total number being 2,000, had embarked in 47 ships which came to an anchor together at the island of Hichi Island, where they said to one another: we have a large number of ships, I am afraid that if we suddenly arrive there, we will be fearful that guards will be shocked and shoot at us. Therefore, they sent Doku and the others to present their intentions for the comingFootnote 15 (Toneri, 1965, p. 379).

The question may present itself consequently: why did the forty-seven vessels not land directly on the coast but opted to anchor at Hichi Island first? The rationale behind this decision may have been to prevent the guards from misinterpreting the large fleet as an invading force and potentially opening fire. These guards were the soldiers assigned to accompany Muzuki in the third year of Tenchi. The return of 1400 men, including not only the emissaries but also Japanese prisoners, and the fleeing of Baekje (Li, 2006), was a positive diplomatic offensive by the Tang Dynasty against Japan. The reason for this was that after the fall of Baekje, the Tang and Silla were at war over the sovereignty of the peninsula. Both sides wanted to bring Japan into their fold, and the Tang Dynasty did so for this purpose.

In December of the tenth year of Tenchi, the Emperor died. The son of Emperor, Otomo, became Prime Minister, and the outlawed nobles of Baekje became bureaucrats as a result. This tilted Japan’s foreign policy in favor of an anti-Silla stance and further eased relations between Tang and Japan. However, with the collapse of the Ungjin CommanderyFootnote 16 and the Jinshin War, this positive trend was broken. During this historical period, there was a link between domestic fortification and foreign policy, and Japan’s relations with the Tang and, indeed, Japan’s relations with East Asia remained in a state of turmoil and tension.

Re-examination of Itojo

Itojo was constructed in AD 756, during the eighth year of Tenpyo-Shoho, about a century after Onojo and Kiijo were built. Itojo is situated on the western slope of Takasu Mountain, east of Itoshima in Fukuoka Prefecture, featuring a stone stronghold at the mountain’s base spanning approximately 2 km in length and is slightly smaller in size than Ono Castle. Kibi no Makibi, a two-time envoy to Tang as a Kentoshi, oversaw the construction of Itojo. He commenced his studies as a foreign scholar in the second year of the Reiki era at the age of 24. He studied the Confucian classics and history of the Tang Dynasty, as well as many arts (Wang and Xia, 1984). In AD 716, Kibi went to Tang for the first time and again in the fourth year of Shoho as a deputy Kentoshi to Tang. He studied the culture of the Tang Dynasty and received high praise. At that time, the only two Japanese students, Kibi no Makibi and Abe Nakamaro, who had an excellent reputation in the Tang Dynasty.

In the sixth year of Shoho (AD 754), he was promoted to Senior Fourth Rank, Lower Grade, and DazainodainiFootnote 17. He recommended the construction of the Itojo, and he was given a mandate to do so (Fujiwara and Sugano, 1991). After his promotion, his priority was to advise on the construction of the castle. Why did he advocate building the castle as the first thing he did upon his return to Japan? One of the reasons lies in the beginning of the Tianbao period, with the Tang rebellion. The Tang Rebellion, in this case, refers to the An-shi Rebellion. In the sixth month of the eighth year of Tenpyo-Soho, the construction of Itojo began, and Kibi no Makibi was assigned to oversee it (Fujiwara and Sugano, 1991). In AD 756, he was ordered to lead the construction of Itojo. On the 10th day of the 12th month of the 2nd year of the Tenpyo-hoji era (AD 758), Ono Tamori, the Japanese envoy to Balhae, reported a rebellion by An Lushan of the Tang Dynasty (Fujiwara and Sugano, 1991). Two years later, Ono Tamori brought back a royal decree ordering the Dazaifu to guard against An Lushan, confirming the suggestion made by Kibi in AD 754 that the castle should be built. In the thirteenth year of Tianbao (AD 754), An Lushan entered the court. Prime Minister Yang Guozhong and the Crown Prince repeatedly told Xuanzong about An’s rebellion, but Xuanzong was so bewitched by An’s rhetoric that he believed in An. In AD 755, An rebelled against the Tang Dynasty and led an army of 200,000 soldiers southward, resulting in the fall of the eastern capital of Luoyang. In AD 756, An exceeded his powers and proclaimed himself Emperor in Luoyang, establishing his own country called Dayan, with his reign known as Xiongwu. The rebels then advanced westward, breaching Tongguan, and subsequently captured Changan, leading to Emperor Xuanzong fleeing to Sichuan.

The An-shi Rebellion shook up the international order in East Asia, with the Tang Dynasty at its center. In AD 756, Xu Guidao sent Zhang Yuanjian, and in AD 758, Wang Xuanzhi sent his general Wang Jinyi to Balhae twice to ask for help, but Balhae refused to send troops because “It was difficult to trust him, and Balhae had left Jinyi behind to send an envoy to ask for more information, but the envoy had not yet arrived and the matter was not known” (Fujiwara and Sugano, 1991). The Balhae Kingdom remained neutral. The Tenno of Japan, upon hearing this, immediately ordered the Dazaifu that “An Lushan was suspected of not being able to attack the west and would also invade the east (refer to Japan)” (Fujiwara and Sugano, 1991). The Japanese side was deeply concerned about An’s rebellion and ordered the Funenoo and Yoshibe Makebi to guard the west gate. Meanwhile, Silla found itself without a strong backer, facing the potential threat of a Japanese expedition to the west and the Balhae to the south.

Itojo was not only in response to the An-shi Rebellion, but it was also a reflection of the delicate relationship between Japan, Silla, and Balhae in the broader context of the international order in East Asia.

Another reason for the construction of Itojo was the joint Japanese-Balhae plan to conquer the Silla. The construction of Itojo was in response to the deterioration of relations with Silla (Mukai, 2016). With the Tang Dynasty in turmoil, the loss of Silla’s patronage provided the perfect opportunity for Japan and Balhae to join forces to conquer Silla. In the six years following the outbreak of the An-shi Rebellion, the two countries exchanged envoys eight times, five dispatched by Japan and three returned by Balhae (Yu, 1995). To conquer Silla, Japan ordered ships to be built throughout the country, including 89 ships for the Hokurikudo countries, 154 ships for the Sanindo countries, 161 ships for the Sanyodo countries, and 105 ships for the Nankaido countries, totaling 500 ships (Yu, 1995). Japan was well prepared for military warfare, and the construction of Itojo was a natural defensive priority.

As Mun of Balhae faded from the plan and the An-shi rebellion came to an end, Japan’s plan to conquer Silla did not take hold, and Itojo did not appear in the war. However, the comparison between Itojo and Onojo, which was built a century earlier, also sheds light on Japan and East Asian relations. The first is the timing of the construction. It took more than a decade to build Itojo, beginning in the eighth year of the Tenpyo-shoho era (AD 756) and ending in the second year of the Jingo-keiun era (AD 768). Onojo was a mountain castle of the same magnitude as Itojo, with a slightly larger city area than Itojo, and it would have taken approximately a similar amount of time to build. According to the Nihongi, Onojo was completed in the fourth year of the Tenchi era (AD 665), so it is assumed that construction of Onojo began more than ten years ago, around AD 650. Sugihara’s (2017) Chronology of the building of the Dazaifu stated that, by examining the annual rings of Sciadopitys verticillata wood excavated from Onojo, it was found that the date of harvesting has been determined to be around AD 650. If this is confirmed, it means that the construction of Japan’s external fortifications did not begin with the defeat at Baekgang but was initiated as early as around AD 650.

This period coincided with the reign of Emperor Kotoku of Japan. According to the Nihongi, starting from the first year of the Taika era (AD 645), Silla sent envoys to Japan every year to pay tribute, and Japan took pride in this relationship and referred to it as the “Reign of Emperor.” However, an intriguing shift occurred during the second year of the Hakuchi era (AD 651). Despite continuing to pay tribute, Silla’s envoys were attired in Tang-style clothing. Why did this pro-Tang phenomenon emerge? Kim Chunchu, the 29th king of Silla (later known as King Taejong Muyeol), played a pivotal role. During Silla’s conflict with Baekje, Kim Chunchu embarked on diplomatic missions to Goguryeo (AD 642) and Japan (AD 647). His goal was to ask neighboring countries to send troops to support him, but he didn’t achieve his goal. According to Samguk Sagi, in AD 648 went to the Tang Dynasty as an emissary to complain to Emperor Taizong of Baekje’s aggression and asked for troops to fight against Baekje, which was agreed by Emperor Taizong. Kim Chunchu then offered to change Silla’s dress code to Tang clothing to submit to the Tang system, and left his son to serve with the Taizong. In the following year, the wearing of Tang clothing began within SillaFootnote 18 (Kim, 1980). That’s why the Silla envoys to Japan wore Tang clothes. During this period, Silla adopted various Tang-inspired policies, including dressing in Tang clothing, adopting Tang-era names, and emulating Tang etiquette and laws. By doing so, Silla signaled its alignment with the Tang-centered order of the world, emphasizing its connection to the powerful Tang Dynasty (Takeuchi, 2023).

The rapprochement of Silla-Tang relations triggered tensions in Japan’s external relations. Various historical records from the same period also show Japan’s hostility towards Silla, such as Minister Kose-no-Oomi even expressed his intention to conquer Silla (Toneri, 1965, p. 317). During this era, there was a shift in the diplomatic dynamics among the three nations, with Silla and Tang adopting a more conciliatory stance while Japan started displaying signs of hostility towards Silla. Hence, this research posits that the fortification activities in Japan were not initiated solely in response to the incident at Baekgang, but rather in anticipation of escalating tensions in foreign affairs, possibly commencing around the year AD 650.

The second is the style of construction. Onojo was built by the Baekje craftsmen Eokrye Bokryu and others and was known as a Joseon-style mountain castle; Itojo was built a century after Onojo and was constructed by the Kentoshi Kibi no Makibi and was known as a Chinese-style mountain castle (Kagamiyama, 1937). It is assumed that he used the techniques he had learned in China to build the castle and that the change from Joseon-style to a Chinese-style mountain castle reflects the exchange between Tang and Japan in the century following the construction of Onojo. The diplomatic exchanges between the Tang Dynasty and Japan during the period spanning from Monmu Tenno’s reign (AD 697–707) to Koken Tenno’s reign (AD 749–758) included the sending of four envoys. Abe no Nakamaro’s persistent stay in Tang and Jianzhen’s multiple crossings into Japan, along with Nakamaro’s appreciation for Tang culture and Jianzhen’s repeated journeys to Japan, are noteworthy anecdotes from this era.

In the latter half of the 9th century, the Korean peninsula entered the Later Three Kingdoms period (AD 889–936), characterized by turbulent times, frequent crop failures, famine, depleted reserves, and civil unrest within the realm. This period saw the rise of Silla pirates, who launched multiple raids on Japan, significantly affecting the diplomatic relations between the two nations. During the 12th year of the Zhenguan era (AD 870), Urabe no Otokusomaro, a native of Tsushima Island, was captured by the Silla kingdom while inadvertently crossing into their territory in pursuit of a bird. While in captivity, he observed the nation engaging in shipbuilding, timber transportation, ceremonial drumming, horn blowing, and military training (Fujiwara et al., 1971). He knew that it was Silla’s preparation for the crusade against Japan’s Tsushima Island. He was arrested by chance, but he was informed of the latest military developments in the Silla Kingdom. He managed to escape back to Japan and reported the incident to the Dazaifu. Japan was not unprepared but prepared to defend itself. The prefectures by the sea were ordered to be fortified, and the country of Inaba, Baji, Izumo, Ishimi, and Oki was ordered to repair their defenses (Fujiwara et al., 1971). Tensions between the two countries are further strained as the Silla army builds ships to conquer Tsushima, and Japan once again builds castles. During the 200-year period between the 8th and 9th centuries, Japan and Silla had strained relations, leading Japan to take a more pragmatic approach towards Silla. In the 8th century, plans were made for a Silla expedition, although they were never carried out. By the 9th century, extensive fortifications were constructed across the country as a defense against foreign threats. Japan also built widespread fortifications after the battle of Baekgang, underscoring Japan’s tense relations with East Asia during the Nara and Heian periods.

The continuation of Japanese fortifications

In AD 1236, Genghis Khan unified the Mongols and started the journey of a great empire. In AD 1259, after 30 years of resistance, the Korean peninsula surrendered to the Mongols. In AD 1262, Kublai came to the throne with his ambitions set on Japan, whose plans were revealed in the letters of state sent by messengers in AD 1274; when the Mongol-Goryeo alliance first conquered Japan, the Yuan army was defeated due to a lack of troops. However, the strength of the Yuan army and the sophistication of their weapons were a great shock to Japan. After the Battle of Bunei, the Shogunate ordered the GokeninFootnote 19 to build a fortification around Hakata Bay to prevent further Mongol attacks, stretching from Osumi Hyuga, Hizen Higo, Chikuzen Chikugo, to Satsuma and Toyo, a total length of 20 km and up to 3 m high. In AD 1281, Kublai gathered 140,000 troops from Goryeo and the Nansong Dynasty to invade Japan again, but they were defeated by a typhoon. Since then, the idea of kamikazeFootnote 20, that the Japanese people are blessed by their gods, has also emerged in Japan after Japan had achieved two unattainable victories against the mighty Mongols.

The Japanese military, whether engaged in domestic or overseas conflicts, demonstrated expertize in self-defense against foreign aggression through the construction of fortifications. This unique capability persisted until the late 16th century during the Japanese invasions of Korea. Upon unifying Japan, Toyotomi Hideyoshi directed his military campaigns toward Korea, establishing a battleground on the Korean peninsula. The well-established defense line of the Japanese army, originating from Japan and extending to the Korean peninsula, played a crucial role in their strategy. The Japanese army’s fortifications on the Korean peninsula were also called WajoFootnote 21. In April AD 1593, the Japanese army retreated from Seoul to the southern coast of the peninsula and built 18 castles between Sosenpo and Uncheon from mid-July to late August. The Japanese invaded Korea again in AD 1597, this time building seven castles on the southern coast of the peninsula (Luo, 2012). The number of fortifications built by the Japanese in Korea is evident, and not only that, but the sites chosen by the Japanese were the same as those in Korea. The post-war fortifications of Baekgang extended from the northern part of Kyushu to the coast of the Seto Inland Sea, and as mentioned above, the Castles were concentrated in the coastal area in terms of the distribution of sites. At the time of the Mongol invasions, the Japanese also concentrated their fortifications at the coastal port of Hakata and built a large number of hill fortifications for defense. Notably, even in the Korean Peninsula, the Japanese defenses continued to follow a similar pattern, hugging the mountains and coastlines in the southern coastal areas. Thus, it can be seen that the selection of sites along the coast and rivers was a significant feature of the Japanese army when building cities.

Conclusion

By examining the construction timelines of Ono Castle and Ito Castle, this study posits that Ono Castle could have been erected as early as circa AD 650, challenging the prevailing scholarly notion that its construction postdated the Battle of Baekgang. This assumption is supported by a range of historical references cited in the text. Specifically, the year AD 650 witnessed a rapprochement between the Silla Kingdom and the Tang Dynasty, leading to heightened intricacies in Japan’s international dealings and a gradual souring of relations with Silla. Therefore, it is proposed in this study that Japan sensed external tensions well before the Baekgang conflict and likely commenced fortification constructions around AD 650.

China and Japan have experienced five significant wars since ancient times: the Battle of Baekgang in the Tang Dynasty, the Battle of Bunei and Koan in the Yuan Dynasty, Japanese invasions of Korea (AD 1592–1598) in the Ming Dynasty, and then the Jiawu War, the Japanese war of aggression against China. Throughout its history, Japan has built fortifications throughout the ages, from Muzuki and stone fortifications at home to Wajo on the Korean peninsula, expanding on the country’s defensive strengths and reflecting the relationship between Japan and East Asia. However, the intricate ties between Japan and East Asia defy simple categorization as positive or negative. Japan’s progression from the decisive loss at the Baekgang battle to the resounding triumph in the Jiawu naval conflict persistently disrupted the regional order in East Asia and scrutinized the acumen of its leaders.